Future of Pedagogy Discovery Group Members
Ashok Goel, School of Interactive Computing, co-chair
Wendy Newstetter, College of Engineering, co-chair
Laurence Jacobs, School of Civil & Environmental Engineering, steering committee liaison
Terry Blum, Institute for Leadership and Entrepreneurship
Julie Kim, College of Design
Pete Ludovice, School of Chemical and Biomolecular Engineering
Michael Schatz, School of Physics
Four Main Findings
The discovery process has led us to four main findings.
1. Educational technologies continue to develop rapidly, given recent advances in computing, networking, media, artificial intelligence, virtual reality, data analytics, etc.
These technologies are starting to significantly supplement human teaching, making many models of learning scalable and repeatable. We should soon be able to support learning for most anyone, anywhere, anytime, and for most any competency.
2. In order for Georgia Tech to stand out as an educational institution given the widening adoption of educational technologies, we urgently need to broaden our educational goals to include whole human development.
This encompasses development of interpersonal and intrapersonal skills in addition to cognitive skills and abilities. Whole human development also entails providing students with a place where they feel comfortable (home), continually challenging and supporting teachers and students alike to learn and grow (edge), and presenting them with opportunities to practice, take risks, fail, receive constructive feedback, and become more proficient (groove).
3. While we can formulate new learning goals that take advantage of the new educational technologies, we also need conceptual frameworks for relating the means to the ends.
A perspective from systems thinking that views pedagogy as a complex system helps relate the new educational technologies to the broader goals of human learning, making important features such as formative assessment salient; a lack of such a framework may result in ad hoc changes to current pedagogy.
4. However, new learning goals, new technologies, and new frameworks are of limited use unless we also build a culture of educational innovation at Georgia Tech.
A culture of education that engages and rewards faculty and students is critical for ensuring that we remain ahead of technological developments and use new educational technologies in the service of human learning; the lack of a culture of educational innovation may result in major, uncontrolled technological disruptions.
New Technologies for Learning
Progress in science and technology continues without pause; in fact, it appears to be accelerating. Thus, we understand the cognitive, social, and cultural processes of learning better than we did a generation ago.
For example, we now have a better understanding of the role of metacognition and the importance of formative assessment and feedback in learning.
Further, the role of context and activity in creating conditions for the transfer of knowledge and skills to different contexts have become more salient. Parallel advances in educational technologies — especially in computing, networking, media, virtual reality, artificial intelligence, machine learning, and data analytics — are making many familiar models of learning scalable and also introducing new models of learning.
Georgia Tech’s Online Master of Science in Computer Science (OMSCS) program illustrates both how new computing technologies make a model of learning scalable and also introduces a new transmission model of learning (although studies have shown that this model may not be appropriate for all learners).
Here we identify a few of the current and future trends important for envisioning 21st century pedagogies.
The apprenticeship model of learning in which the learner gains skills and knowledge through one-on-one or small group interactions with a teacher has been around since Socrates.
This model of extensive teacher-student interaction enables the learner to progress at a pace and in a style suited to him or her through the process of personalized scaffolding. Bloom (1984) argued that the average student who received one-on-one tutoring from an expert tutor scored two standard deviations higher on standardized achievement tests than an average student who received traditional group-based instruction.
While it is a model that works for almost any competency, with the requirements of mass education, it has become rare in postsecondary education (except for research labs) because the model is neither easily scalable nor financially affordable. Thus, as the class sizes became bigger, the instruction becomes more uniform and impersonal.
With class sizes at Georgia Tech reaching as many as a few hundred students (even for some graduate classes), the interaction between the teacher and the students is so limited that a typical teacher may not know the names of all her students. As a result, students can feel disengaged from the learning.
The Holy Grail, then, for future pedagogies is to find ways to personalize instruction for all students. In an interesting twist, technology now appears to be catching up with the requirements of mass education, making personalized learning increasingly feasible and affordable again.
Advances in artificial intelligence and machine learning bring with them the potential to mimic the personalized interactions, the real-time data collection and interpretation, and the pedagogical responses found in one-on-one learning interactions. (Georgia Tech’s Jill Watson project is only one point on this rapidly advancing envelope.) Together these advances suggest that Stephenson’s (1995) vision of “A Young Lady’s Illustrated Primer” might not just be some futuristic science fiction anymore.
Intelligent Tutoring Systems (ITS) can modify the presentation and sequence of materials in response to student performance.
The most effective systems capture fine-grained data behind the scenes and use learning analytics to enable personalized responses, thereby serving as a tutor to the student. Cognitive tutors — or more accurately, problem-solving and solution-analysis tutors — assist students in developing procedural, rule-driven skills, especially in math.
Cognitive tutors are predicated on the ability to provide just-in-time remedial feedback and decide when to move on to a new topic.
In contrast, content sequencing tutors seek to tailor the learning content or learning pathway based on an accurate assessment of a large array of skills with the least possible amount of evidence.
More recently, example-tracing tutors (Aleven, McLaren, Sewell & Koedinger, 2016) evaluate student behavior by flexibly comparing it to examples of correct and incorrect problem-solving behaviors.
These tutors are capable of sophisticated tutoring behaviors in providing step-by-step guidance on complex problems while recognizing multiple student strategies and maintaining multiple interpretations of student behavior. Authoring tools for these systems have reduced the learning curve and time required to tailor the system to a course.
While these systems target cognitive competencies, other IT systems are addressing the intrapersonal competencies. The Self-Assessment Tutor, an intelligent tutoring system for improving the accuracy of student judgments regarding their own knowledge, focuses on self-regulation and metacognitive skills.
Another such system (Bernacki., Nokes-Malach, and Aleven, 2013) collects data to assess motivational variables and examine their association with cognitive and metacognitive behaviors for students learning mathematics.
Another research and development area that holds promise is virtual tutors. These animated pedagogical agents are lifelike virtual characters designed to enhance learning. Artificial tutors are being developed with the ability to perceive emotions experienced by learners, and to incorporate these into pedagogical strategies.
The presence of a tutor, embodied as a 2D or 3D character, has shown some positive learning effects, particularly in student engagement.
Methodologies that extract useful and actionable information from large datasets has transformed one field of scientific inquiry after another.
When applied to education, these methodologies are known as learning analytics (LA) and educational data mining (EDM). Large educational datasets present opportunities to discover patterns that occur in only small numbers of students or frequently, to develop an understanding of student uses of different learning resources and different outcomes, to undertake fine-grained analysis of phenomena that occur over long periods of time (e.g. disengagement) and to illuminate the impact of variables of interest in a learning environment.
More recently, datasets can also provide a window on the quantity and quality of online interactions in learning collaborative settings. The availability of such data has proven to be of great importance in promoting retention and completion as evidenced by the GSU Predictive Analytics Support system.
Well-designed data collection strategies and dashboard have the potential to provide real-time data on individual students and groups that can both enhance learning and prevent failure.
Technology now offers unique opportunities to:
- Simulate a physical presence in the real world or in an imagined world,
- Overlay content on the real world, and
- Merge real and virtual worlds to produce new environments and visualizations where physical and digital objects co-exist and interact in real time.
These three represent virtual reality (VR), augmented reality (AR), and mixed reality (MR) scenarios. Each has potential application in education in providing new learning modalities. For example, consider Microsoft’s Holodeck currently being used in medical education as an illustration of educational VR technology.
At the same time, online simulations of workplaces can also be seen as a virtual reality of a different kind, where a learner can experience tasks that might be found on the job, but in an educational setting. The recent astonishing success of Pokemon Go signals the advent of mixed reality and gaming environments that could be leveraged for learning purposes.
Whole Human Development and Flourishing
Here is one approach to deep cultural change that supports educational revolutions: Draw on research-based principles from developmental psychology to make Georgia Tech a Deliberately Developmental Institution (DDI), i.e., an institution whose central focus is whole person development. Imagine a Georgia Tech that offers the following to the rising high school senior, to the college senior considering graduate study, or to the working professional seeking to grow her skill set:
When you ask, “Why should I pick your institution?,” Georgia Tech whole-heartedly responds, “We want you to join us because we will focus on helping you flourish, to thrive, to become a better ‘you.’”
These are not empty words. At Georgia Tech, world-class educational experiences in your chosen fields of study are systematically designed using principles grounded in solid scientific research on human development and learning.
Your Tech experiences aim to help you acquire deep knowledge in your subject woven together with skills you need to adaptively apply that knowledge to overcome challenges in your career and your life — those skills are not only cognitive (e.g., problem solving, critical thinking, etc. ), but also interpersonal (e.g., teamwork, leadership, etc.) and intrapersonal (e.g., the ability to reflect on your process of learning to help you learn more effectively).
To make this happen, Georgia Tech draws upon cutting-edge pedagogies, novel learning technologies, and optimized spaces (real and virtual); moreover, as we move together along the different stages of your learning, we provide feedback to help you advance and we learn from you ways to improve your learning experiences.
At every stage and for every experience, you are immersed in an environment that will push you continually to confront your limits — that’s going to be uncomfortable and make you feel vulnerable — and that will be OK because at Tech we will hold your vulnerabilities well and surround you with supportive practices that will enable you to overcome your limits.
This is how we do learning at Georgia Tech because the best available science on learning and human development tells us that we all must learn in this way in order to grow and thrive and become more fully ourselves, to become the person you want to be.
Two key ideas underpin this vision of future educational experiences.
(1). Multidimensional Character of Learning Experiences
Traditionally, most educational experiences in a university setting focus on supporting learning of content in a particular subject domain (physics, chemistry, engineering, etc.). A more modern view of such experiences recognizes that learning inherently has a multidimensional character (Figure 1).
The Cognitive dimension commonly includes competencies such as innovation, critical thinking, problem solving, information literacy, reasoning, and argumentation. The Intrapersonal dimension recognizes the importance of skills such as flexibility, initiative, appreciation for diversity, and metacognition (the ability to reflect on one’s own learning and to make adjustments accordingly).
The Interpersonal domain includes abilities such as communication, collaboration, responsibility, and conflict resolution.
In traditional STEM learning, the importance of cognitive competencies is widely recognized, but rarely supported explicitly; STEM instructors often assume that students will “pick up” cognitive skills as they learn new subject content. It is rare that intra- and interpersonal competencies are accounted for in learning at Georgia Tech.
Future educational experiences must be designed to account explicitly for all these learning dimensions; Georgia Tech graduates, and others, plateau or flame out in organizations not because of their cognitive limitations.
While IQ and cognitive competencies are necessary, they are not sufficient, positive psychological capital (i.e. the combination of hope, resilience, optimism, and efficacy); emotional and social competencies are the differentiators and are fostered by nurturing intra- and interpersonal skills.
Figure 1. Designing excellent educational experiences in a particular subject domain requires explicit consideration of cognitive, intrapersonal and interpersonal competencies. (Figure courtesy of Jim Pellegrino)
(2). The Culture of Whole Person Development
Advances in the science of developmental psychology suggest that a strong culture of Deliberate Development (of both an institution and the individuals within the institution) requires three interlocking pillars (Figure 2). To develop, individuals must be placed on “Edge” through revealing their personal limitations and taking risks to overcome these limitations. Exposing weaknesses for the purpose of growth can only happen in a supportive environment (“Home”) where each person’s vulnerabilities are well-held. At the same time, the institution must support regular, continual practices and routines (“Groove”) that enable individuals to continually work on overcoming their limits and to experience growth.
Educational institutions like Georgia Tech are in the people development business; however, the culture of such institutions typically contains serious flaws that are major impediments to Deliberate Development. As one example, the culture of Georgia Tech currently fails to provide a safe “Home” where students can feel supported in revealing their weaknesses and in taking the risks needed for growth. Georgia Tech must examine how to make the culture changes to support “human flourishing” (a central theme in Hoadley’s characterization of WAVE 5 educational advancement) that is necessary for both Tech individuals and Tech as an institution to adaptively meet the 21st century challenges that are increasingly technical, complex, volatile, uncertain, and ambiguous.
Figure 2. Excellent educational experiences require a culture where individuals uncover their weaknesses (identify their Edge) in a safe environment (Home) where regular practices and routines (Groove) enable individuals to grow and to thrive. (Figure from the website: http://www.waytogrowinc.com.)
Viewing Pedagogy as a System
While we can formulate new pedagogical goals and while we understand some of the affordances of new educational technologies, relating the goals to the technologies is complicated: how do we package and use specific technologies to accomplish particular learning goals?
It is important to clearly and precisely specify this connection because otherwise any learning process we design likely will be ad hoc.
Consider, for example. the issue of formative assessment and feedback; while we understand their importance, without specifying the relationship between pedagogical goals and educational materials and technologies, it is hard to specify what precisely to assess, how to assess it, when to conduct the assessment, or what feedback to provide, how and when.
Having a model that accounts for the states in which the instructor can intervene and the interactions between these states is actually essential to ensuring the design of appropriate innovative educational strategies.
Given any system — natural, social, or technological — humans in general understand the visible structure, components, and connections of the system and have some comprehension of the system’s abstract functions or goals, but show minimal (often close to zero) understanding of the system’s invisible behaviors, processes, and mechanisms – how the system actually works.
This is also true of pedagogy as a system. Both teachers and students alike typically understand the visible structures, such as the physical spaces, the educational materials, the technological tools of teaching and learning; both teachers and students also have some notion of the intended abstract learning goals and outcomes of an educational course or program.
But neither teachers nor students manifest deep comprehension of the invisible processes and mechanisms that map the visible structures into the abstract goals, that arise out of the interactions among structural elements and result in the accomplishment of the goals. In fact, very few faculty in the design of their courses even think about this behavioral level or the various states changes that need to occur within the student that constitute learning.
Nor do they think about the assessment strategies that they need to employ to assess the degree of change that has occurred.
However, without this comprehension or practice of designing with behaviors in mind, two negative outcomes can occur. First, the intended learning outcomes may not be achieved in the majority of the students because the processes by which such learning could occur had not been designed into the course. Second, neither the teacher nor the student would understand the significant role of assessment and feedback in the learning process.
Figure 3: The Structure-Behavior-Function model of a pedagogical design relates various pedagogical elements and processes so that the invisible learning processes (the behaviors in the figure) become visible and perspicous and the implicit role of formative assessment and feedback becomes explicit and salient.
Structure-Behavior-Function modeling is a framework for potentially making the invisible processes of successful pedagogogical design visible to teachers and students alike.
The model comprises three interacting levels. The top level specifies the intended learning goals and outcomes of the pedogical designs.
The middle or behavioral level illuminates the invisible processes of state changes within the learner as well as the interactions and activities indicated or mecahmisms that cause the change to occur. Further formative assessment provides data about the nature of the state change.
Finally, the structural elements of the class — the materials, the space, and the technologies — are a kind of bedrock or foundation for the two levels above. The SBF model explicitly specifies the relationships between the structures and the behaviors as well as the behaviors and the functions, so that if a formative assesement is negative, the teacher has some diagnostic knowledge about what feedback to give and how to revise the materials to accomplish the learning goals.
Culture of Educational Innovation
Sophisticated knowledge and skills in the right hands and minds can empower students and educators to make smarter and different choices. These tools can open up new and exciting possibilities for the direction of education, methodology, and practice.
However, these instruments are not prescriptive, so we must seek the balance between the knowledge and skills with the sensibility and minds to use them. We all aspire to be educated users with visionary dreams.
Georgia Tech is uniquely positioned to foster a culture of educational innovation defined by robust models of collaborative engagement, creative problem solving, and critical thinking.
Educators need to not present themselves as the experts, but as the guide to enable students to devlelop their own expertise. Key to this is a philosophical shift in the educator to ask, not tell, the student what they’ve learned, created, and gained from a shift in their perspective. Our role as educators shifts to one of designing the process and experience of learning rather than delivering information, e.g. active versus passive learning.
Educator-designed experiences yield critical thinking in the classroom as students are coached to think through problem-solving like experts.
In this interactive environment, learning is reciprocal, where educators can learn just as much from their students. Technological advancements have transformed how we teach and learn, where the educator creates and crafts the learning vision, using technology and teaching spaces to deliver that vision.
When learning is less constrained by physical space and a timed class schedule, the space of the classroom can morph into an environment where problems are solved and assumptions and misconceptions are challenged.
Educators can develop constructed situations for students to generate and test their knowledge, to conduct research, and to attempt, fail, and try again.
We redefine the student perspective of success as a series of failures from which they redesign their ideas to test again. Fundamentally, this is the underpinning of collaborative engagement, creative problem solving, and critical thinking, the whole person development that successful graduates will require in their future careers.
We sit at a significant juncture where, because of the technology, we are enabled to build a robust model that challenges the nature of education, both present and future. This will require taking risks (being on the edge) with the support from the Institute that fosters motivation and nurtures an environment of curiosity (being in the groove).
To build a culture of educational innovation will take time and commitment from students, faculty, and the administration.
Georgia Tech has an opportunity to bring together intellectual depth and curiosity, research, and creative production of our community of scholars to establish the cultural benchmark for education innovation.
Implications for Georgia Tech
Characterize areas where significant advances might be made over the next five years, ten years, and beyond.
The near term, five to ten years, should see a wider adaption of pedagogical approaches that embrace student-centered learning techniques where the student takes a more active role in the education process. The advances here should come in taking these techniques to scale and enabling larger numbers of students to participate.
An example of this would be the Vertically Integrated Projects (VIP) program that unites undergraduate education and faculty research in a team-based context. Undergraduate VIP students earn academic credits, while faculty and graduate students benefit from the design/discovery efforts of their teams.
VIP extends the academic design experience beyond a single semester, with students participating for up to three years. It provides the time and context to learn and practice professional skills, to make substantial contributions, and experience different roles on large multidisciplinary design/discovery teams. The long-term nature of VIP creates an environment of mentorship, with faculty and graduate students mentoring teams, experienced students mentoring new members, and students moving into leadership roles as others graduate.
In each time frame, construct scenarios that would fundamentally alter Georgia Tech’s approach to instruction and learning.
At present, the dominant STEM pedagogy at Tech and elsewhere with certain exceptions, is teacher-driven lecturing with a smattering of active learning.
The focus is on the acquisition of cognitive skills and knowledge, with little or no emphasis on the other two dimensions. There are some notable exceptions, however. Problem-based learning (PBL), the pedagogical approach foundational to the biomedical engineering curriculum, focuses on all three dimensions by having teams of students take on authentic, complex, ill-structured, and ill-constrained problems early in the curriculum.
Teamwork and collaboration, leadership, communication, and conflict resolution are essential skills practiced on the PBL teams as they navigate and negotiate the problems. Further, tolerance for failure, reflection, and a developing sense of self-efficacy are potential outcomes of this learning experience. The problem-solving studio, an offspring of PBL and also practiced in BME, focuses on the development of grit, an important intrapersonal trait.
Maker spaces learning activities, both personal and class-driven, are interesting in their potential to address both the cognitive and the intrapersonal.
These “practicum” innovations and complementary assets are integrated into the Scheller College of Business curriculum. These spill over into STEM in interdisciplinary programs such as the Leadership minor, the Technology and Management minor, the TI:GER program, and the entrepreneurship certificates that the college offers for non-business students.
What steps can Georgia Tech take today to anticipate scientific or technological advances?
Faculty who are versatile with anticipatory anthropology, futures studies, etc. could contribute to this initiative. Specifically, we can look at technology below or just above the horizon and begin to brainstorm applications.
For example, the Pokémon Go phenomenon seems like a wild new thing today, but many of the leading critical thinkers in education and anticipatory studies have seen this coming since the early 1990s.
The discussion on augmented reality raged back then — it just took 25 years to develop the actual hardware to pull it off. Another possible methodology would be to select a new tech that just came online and test what its impact may be in education. Again, the Pokémon Go phenomenon could be a great opportunity to embed pedagogy around, say, environments for students and, eventually, clients.
There is another possible path. We can consider the remarkable pace of technological change and innovation as well as scientific discoveries (often in relation to one another). We can look back to history and see the curves of early adoption of these initial technological or scientific breakthroughs. From this perspective, we can study how we actually used, engaged, and appropriated the software side of things. In other words, focus more directly on the how versus what. It is critical to develop the cognitive dimension, but we must recognize that “cognitive” has the potential weakness of emphasizing the neocortex, particularly the frontal cortex of our brain where ego, criticism, logic, and the intellect dwell. All important and necessary but, as Susan Sontag stated, we run the risk of furthering the hypertrophy of the intellect from which we all suffer. We must embrace ethical, empathic, and aesthetic wisdom. By critically reviewing the past, we can make projections to the future and anticipate advances in science and technology.
A third possible path brings the two ideas together by looking backward to allow us to look forward. Interesting technology/scientific developments to watch with direct effect on education includes internet of things, drones/robots, quantum computing, AI, genetic engineering, H-C interface, etc.
Given the importance of creating an ongoing capacity for early identification and implementation of collateral learning assets above core disciplinary cognitive knowledge, a forward-thinking unit — perhaps an Office of Pedagogical Innovation (OPI) charged with identification, enhancement, and dissemination — should be established.
Perhaps it should be within the Center for Learning and Teaching (CTL) and become the repository of potential innovations, pulling innovation from faculty, running experiments, and helping faculty easily implement them. The OPI should scan the environment for emerging ideas, accept nominations of ideas, recombine different opportunities, write white papers, and otherwise create a strategy for anticipating scientific and technological advances.
Another structural change could include an academic unit that is a degree-providing unit in pedagogical innovation and design. Our culture respects areas where new knowledge and research are created along with students who specialize in the disciplinary area.
Bringing together those across our current disciplines who work in cognitive areas and augmenting them with others who work on other areas of the CI2 (interpersonal and intrapersonal capabilities), both those currently at Tech and those who would be added, would create a permanent structural element that has the components of other academic units and the stature to prepare students to lead the educational challenges of the future beyond Georgia Tech. This unit would build upon adjacencies to understand and serve diverse learners.
Given our goals, structural changes could provide us with the opportunity to be permanently, deliberatively, and innovatively developmental in our teaching and learning. In any event, the new structures and strategies should integrate new practices into the academic units. Like the rest of the Institute, the entrepreneurial spirit will need to be allowed to flourish with the nourishment of tangible and intangible resources.
Clarity of vision, celebrating the innovations and innovators, enabling innovators, and having exemplars who are respected among their peers will be essential in changing the culture.
We cannot stop with the providers and purveyors of innovation. We need to prepare our students for a different type of learning. They will continue to come from K-12 systems that use old, tired techniques. We must focus on how we motivate them to learn in new ways with new rules.
What resources or other missing elements would need to be provided to take advantage of positive advances?
“It’s hard to make predictions, especially about the future.” Making good predictions in the CNE context depends crucially on how we view the challenges/opportunities that Georgia Tech education will face in the future.
One starting point (following the work of Ronald Heifetz) is to ask the question: To what extent will meeting the challenges/taking advantage of the opportunities require advances that are primarily technical or primarily adaptive?
It’s in Georgia Tech’s DNA to focus on the technical — the interweaving of new technologies (VR, augmented reality, enhancements using big-data, etc.), spaces (virtual and real), pedagogies, and assessments; the report on the future of MIT education reflects an effort to peer into the educational future from a primarily technical stance.
However, robust construction of technical roadmaps for Tech’s educational future depends on key adaptive advances in the continual formulation/reformulation of principles and mindsets (both individually and institutionally) that are needed for proactively and flexibly responding to challenges/opportunities, especially when those challenges/opportunities are particularly difficult to foresee far in advance.
Pursuing technical advances alone (greater integration of smartphones into educational experiences, as just one example) without adaptive advances (which, in essence, are changes in culture) increases the risk of unintended adverse impacts (e.g., the educational equivalent of texting while driving).
Buy-in by campus leadership is crucially important for nurturing cultural shifts and adaptive thinking needed to guide change in the near-, mid-, and long term. Campus leaders need both to demonstrate the willingness to do the necessary interior work (both individually and institutionally) and to serve as models for adaptive change.
One approach to leadership buy-in can begin in the CNE Ideation phase: Assemble willing campus leaders in groups focused on particular Tech learner constituencies (e.g., undergraduate learning, graduate learning, professional education, …) to grapple with questions of culture that impact educational change, for example (a) How to shift the culture in the face of legacies? (b) How to deal with power and power shifts? (c) How to face the inevitable coup attempts?
These groups would need to create shared norms and “rules of the road” that support a safe, supportive “container” (a “home”) to engage deeply with questions of culture and culture change.
This cultural Ideation should draw up local expertise for support; for example, (1) Ideation efforts are, in effect, “startups”; thus, leadership groups should employ “Flashpoint” methods, in consultation with Merrick Furst and collaborators, (2) Leadership groups would benefit from consultation with Atlanta-based Way to Grow Inc., headed by Andy Fleming, an expert on Deliberately Developmental Organizations.